Fix multiple deletions of the same files in save_pretrained #16947
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Checkpoint sharding introduced a bug in
save_pretrained
for distributed setups, where the function is called on every process (TPU training for instance, or the scripts with Accelerate (see this issue).This changes the logic to only remove files when:
Except
save_pretrained
does not know if we hare on process zero or not, so I usesave_config
to detect that. It looks like that argument was not aptly named and should beis_main_process
instead. I can go ahead and deprecate/rename in this PR if you agree.